The method uses a contrarian technique to manage the AP channel settings in a way that would cause many of us to shudder. It is easily explained on a whiteboard, harder in writing - but let me have a go. In a "normal" multi-AP deployment I call it "normal" because it is the established and successful technique as used by the largest and most successful enterprise Wi-Fi vendors , the APs as much as possible select different 20MHz "channels" - and sometimes multiple "bonded" 20MHz channels - within their defined 2.
By being on a different channel, it reduces the interference between APs which might otherwise negatively impact performance and overall network throughput. The good news, by using this "normal" deployment method you get loads of bandwidth. If you have 10 APs and each has a theoretical 1Gbps of throughput assuming it was an In this "normal" model, each client device has a direct "association" with a specific AP until it drops that connection and makes an association with another AP.
If you roam across sub-nets the glitch can be greater the video might stutter or more likely a VoIP call might get dropped. This is because the client actually has to re-authenticate and be issued a new IP address on the new sub-net. Some vendors offer tunneling solutions so latency sensitive apps especially voice always get tunneled back to a centralized controller so that as you roam, even across sub-nets, the client IP address is unchanged so calls do not drop.
Ruckus, Aruba and others offer such tunneling options. Now lets look at the Zero Hand Off type solution. I'll call that solution "deviant" as opposed to "normal" well, I am a competitor to that type of deployment technique after all so what would you expect?
So in the deviant model the central controller assigns every AP in the network to the SAME channel instead of different channels. On paper this would suggest that massive inter-AP interference should occur. However, such interference does not occur because in fact the controller only allows one AP in the entire "seamless" network to transmit at a time.
In theory that AP suffers zero interference because the other APs are all forced to remain quiet. In this deviant model, each client is actually not "associated" to any particular AP The controller itself then makes the decision about which physical AP to transmit or receive data to the client on.
The client never knows which physical AP is sending the data or picking up its responses - it does not need to know and it seems to just magically roam transparently across the network. Well all that seems just wonderful doesn't it? But as my Dad always told me You see, because each physical AP has to be on the same channel and can only transmit when instructed and when every other AP is instructed not to transmit it means the entire network capacity is the equivalent of just one AP.
Yes you have 10 APs, each of which can theoretically transmit at 1Gbps As mentioned before, the simple way to think if it is that the controller itself is the AP So now if you have 10 physical APs deployed of 1Gbps capacity each, your entire network capacity is actually still only 1Gbps. Imagine if one location of your network suffers from some really bad transient outside interference something which happens pretty much everyday in most networks btw. In the "normal" deployment, the controller or in the case of Xclaim, the AP itself simply rescans and moves the AP to a new channel with less interference.
The locally connected clients are affected but it is only the clients on that one AP. In the "deviant" model, one of two things can happen. Now remember, in the deviant model, every AP has to be on the same channel for this seamless roaming to work. So, either the AP in the area of interference stays on the same channel and its ability to function is severely restricted so crappy Wi-Fi in that area of your network - sorry about that , or the entire network of APs has to switch to a new channel and every client on every AP is thrown off the network and has to re-associate Extricom have never been able to make it public having been around for a similar length of time, and the vendor offering its version called "Zero Hand Off" gets consistently poor reviews for its performance.
The reality is that deviant Wi-Fi is a good story to spin, it sounds credible and in a world where vendors look for a marketing differentiation, it is certainly a "different" story to spin and claim that the difference is something better.
But it is clear that real-world usability is simply not at all great - and hence the many poor reviews. And lets face it Great writeup, I would summarize it as "bad client behavior in switching to the best available AP results in mass stupidity from a narrow range of vendors.
It is incumbent upon you to look up the reference and read the source and worthwhile, as well. Brand Representative for Xclaim Wireless. Awesome article. I remember reading quite a bit about Meru pushing that model.
We have Ruckus, and a Fortigate. This is great info! I wish we could get more content like this. Easy to read, simple an informative. I kinda knew something was up when one of my clients never heard from Meru and Meru seemed to be afraid to deal with me since I also do benchmarks and testing. I gave the info to the client and the client said they contacted Meru and Meru never contacted my client back.
They prolly knew they wouldn't do good so they bailed on both me and the client. Oh dear I am not sure which is worse Let me know if you want to know why, and why Xirrus have thus decided to get into the standard dual-radio 2. Well How I deployed the Xirrus arrays I had them on different channels like the "standard" setup you have listed above. There were 4 arrays XR It's still workin out great for them, however. Very good write up and simple explanation.
I will be referring to this probably verbatim for customer's comprehensive benefit of the technology at some point in the future, I guarantee it. If AP1 and AP10 are hundreds of feet from each other, there is no reason they can't both transmit at the same time on the same channel. You are commenting using your WordPress. You are commenting using your Google account.
You are commenting using your Twitter account. You are commenting using your Facebook account. Notify me of new comments via email. Notify me of new posts via email.
Test Bed: If you would like to play around with Fortinet-SCA more just setup your test environment for the following and you can come up with your test cases and values. Like this: Like Loading Leave a Reply Cancel reply Enter your comment here Fill in your details below or click an icon to log in:. Email required Address never made public. The cafeteria in this high school along with all other high schools in the district qualify as high density areas and are outfitted accordingly.
My man Rowell Dionicio explains this issue well in a recent blog post at Network Computing. At the time I did not capture any further information so for the sake of this blog post I re-created the scenario at my house.
There were a total of seven APs which suffered from this bug and a few of them were within hearing distance of where this high concentration of clients was located. I checked the controller and found the channel utilization through the roof in this area.
I removed the corrupted APs from the controller and added them back. You did read that correctly…ALL. Both virtual cells were back in business.
Brain matter on the wall is unbecoming. I digress. You may argue that this would be simply a one off issue due to the AP radio corruption that took place. In fact, it is not. It would be very easy to have a problem like this occur with the APs working as they should in an improperly designed network. To properly design a Meru single channel wireless network you should group like model APs in their own virtual cell when in a mixed model environment.
They can all happily co-exist in the same ESS profile though. All the same RF fundamentals, designing, deploying, etc are relevant when using single channel architecture. Poor marketing among other things give it a bad name. Taking it out of the box, racking the equipment, throwing everything on the same channel, and letting it rip WILL yield poor performance. I have read or heard that garbage before. So please…I ask from my little single channel heart, give me and my precious FortiRu a chance.
PS — There will be a Whiskey and Wireless podcast in the near future where I talk about my single channel adventures. Keep an eye on the Whiskey and Wireless Twitter feed.
0コメント